PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise

Advanced search

Message boards : Twin Prime Search : Primegrid hogging processors despite its allocated share

Author Message
Profile Anthony Q. Bachler
Send message
Joined: 9 Aug 07
Posts: 2
ID: 10601
Credit: 14,334
RAC: 0

Message 6776 - Posted: 29 Aug 2007 | 22:56:56 UTC

Ive allolcated Primegrid 1% of my CPU power, so that it runs when my primary project isnt handing out work. The problem is that Primegrid is hogging the CPU and about to get detached because it gives such short report deadlines that nothign else will run untilthe last minute, which then makes my other project end up dumping a bunch of WU by being late.

Anyone have a solution to this problem? I realyl dotn want to have to detach from primegrid, but I will if I cant get it to 'play fair'.

popandbobProject donor
Send message
Joined: 19 Aug 05
Posts: 109
ID: 463
Credit: 3,810,444
RAC: 0
321 LLR Bronze: Earned 10,000 credits (13,319)Woodall LLR Bronze: Earned 10,000 credits (10,490)321 Sieve Bronze: Earned 10,000 credits (25,838)Cullen/Woodall Sieve (suspended) Bronze: Earned 10,000 credits (30,418)PPS Sieve Ruby: Earned 2,000,000 credits (3,072,140)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Bronze: Earned 10,000 credits (27,478)GFN Bronze: Earned 10,000 credits (30,942)PSA Gold: Earned 500,000 credits (574,396)
Message 6777 - Posted: 29 Aug 2007 | 23:29:08 UTC

Sounds like you want primegrid as a backup right?
Suggestion:

Set the resorce share to something stupidly low (ie. 1e-20 ) and keep a small wu' catch. This should solve the problem.

~BoB
____________

Rytis
Volunteer moderator
Project administrator
Avatar
Send message
Joined: 22 Jun 05
Posts: 2648
ID: 1
Credit: 21,592,315
RAC: 2,317
321 LLR Silver: Earned 100,000 credits (104,475)Cullen LLR Silver: Earned 100,000 credits (291,372)ESP LLR Bronze: Earned 10,000 credits (18,156)Generalized Cullen/Woodall LLR Bronze: Earned 10,000 credits (15,259)PPS LLR Silver: Earned 100,000 credits (113,978)PSP LLR Silver: Earned 100,000 credits (116,517)SoB LLR Silver: Earned 100,000 credits (282,655)SR5 LLR Bronze: Earned 10,000 credits (14,071)SGS LLR Silver: Earned 100,000 credits (100,082)TPS LLR (retired) Silver: Earned 100,000 credits (111,607)TRP LLR Jade: Earned 10,000,000 credits (16,076,365)Woodall LLR Silver: Earned 100,000 credits (101,463)321 Sieve Silver: Earned 100,000 credits (201,501)Cullen/Woodall Sieve (suspended) Silver: Earned 100,000 credits (214,653)Generalized Cullen/Woodall Sieve (suspended) Bronze: Earned 10,000 credits (14,200)PPS Sieve Silver: Earned 100,000 credits (363,095)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Silver: Earned 100,000 credits (200,232)TRP Sieve (suspended) Ruby: Earned 2,000,000 credits (2,453,872)AP 26/27 Silver: Earned 100,000 credits (473,058)GFN Silver: Earned 100,000 credits (184,089)PSA Bronze: Earned 10,000 credits (97,541)
Message 6779 - Posted: 30 Aug 2007 | 5:47:47 UTC

BOINC tries to maintain resource share in the long run, not in the short run. It might run PG workunit first until completion, but then it will notice that PG has used more CPU time than it should have, according to resource share, and will not ask for more work for quite a long time (or until work is needed because your main project is out of work).
____________

Brian D from GeorgiaProject donor
Avatar
Send message
Joined: 3 Sep 05
Posts: 78
ID: 612
Credit: 4,568,813
RAC: 0
321 LLR Bronze: Earned 10,000 credits (30,386)Cullen LLR Bronze: Earned 10,000 credits (47,849)ESP LLR Bronze: Earned 10,000 credits (21,583)PPS LLR Amethyst: Earned 1,000,000 credits (1,343,978)PSP LLR Bronze: Earned 10,000 credits (12,472)SoB LLR Bronze: Earned 10,000 credits (15,884)SGS LLR Gold: Earned 500,000 credits (926,215)TPS LLR (retired) Gold: Earned 500,000 credits (763,181)TRP LLR Bronze: Earned 10,000 credits (74,948)Woodall LLR Bronze: Earned 10,000 credits (36,063)321 Sieve Bronze: Earned 10,000 credits (84,038)Cullen/Woodall Sieve (suspended) Bronze: Earned 10,000 credits (62,153)PPS Sieve Silver: Earned 100,000 credits (174,709)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Gold: Earned 500,000 credits (968,173)
Message 6781 - Posted: 30 Aug 2007 | 14:42:28 UTC - in response to Message 6776.
Last modified: 30 Aug 2007 | 14:45:09 UTC

Ive allolcated Primegrid 1% of my CPU power, so that it runs when my primary project isnt handing out work. The problem is that Primegrid is hogging the CPU and about to get detached because it gives such short report deadlines that nothign else will run untilthe last minute, which then makes my other project end up dumping a bunch of WU by being late.

Anyone have a solution to this problem? I realyl dotn want to have to detach from primegrid, but I will if I cant get it to 'play fair'.


What PrimeGrid projects are you running? PrimeGen, TPS, Woo or Cul? Each of these have their own deadlines and varying work unit length. Since you posted under the TPS section, I will assume that you have chosen TPS. TPS work is sometimes sent out as verifications or for unreturned work and the normal 4 day deadline is shortened to 12 hours. Maybe you picked up a batch of these that forced your system into panic mode to crunch PG excluseively.

As Rytis points out, give the projects a few days to run and the BOINC client resource share debt scheduling should solve your problem automatically.

Profile Anthony Q. Bachler
Send message
Joined: 9 Aug 07
Posts: 2
ID: 10601
Credit: 14,334
RAC: 0

Message 6783 - Posted: 30 Aug 2007 | 17:13:29 UTC

The problem is, that if I set the alloca to 1E-11 then the server wont give me any work at all

mdettweiler
Avatar
Send message
Joined: 25 Apr 07
Posts: 271
ID: 7489
Credit: 7,359,328
RAC: 718
321 LLR Bronze: Earned 10,000 credits (10,279)Cullen LLR Bronze: Earned 10,000 credits (11,460)PPS LLR Bronze: Earned 10,000 credits (14,949)PSP LLR Bronze: Earned 10,000 credits (11,016)SoB LLR Bronze: Earned 10,000 credits (16,635)SGS LLR Bronze: Earned 10,000 credits (11,081)TRP LLR Bronze: Earned 10,000 credits (10,140)Woodall LLR Bronze: Earned 10,000 credits (10,331)321 Sieve Bronze: Earned 10,000 credits (20,883)Cullen/Woodall Sieve (suspended) Bronze: Earned 10,000 credits (21,464)PPS Sieve Bronze: Earned 10,000 credits (57,241)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Bronze: Earned 10,000 credits (20,499)TRP Sieve (suspended) Bronze: Earned 10,000 credits (20,039)PSA Turquoise: Earned 5,000,000 credits (7,121,381)
Message 6784 - Posted: 30 Aug 2007 | 17:56:59 UTC - in response to Message 6783.

The problem is, that if I set the alloca to 1E-11 then the server wont give me any work at all

It won't now, but it will when the primary project doesn't have any work to give out.
____________

Message boards : Twin Prime Search : Primegrid hogging processors despite its allocated share

[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2020 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 1.46, 1.41, 1.45
Generated 10 Jul 2020 | 23:51:14 UTC